The Oslo Tomosynthesis Screening Trial
Digital breast tomosynthesis (DBT) is a promising new technology that acquires 3-D breast images. The individual images are presented as thin, high-resolution slices, which can be displayed individually or in a dynamic cine mode. Preliminary studies in a clinical setting have demonstrated that this new technology has the potential to improve not only the breast cancer detection but also to reduce the number of false positives.
Many experts believe that the greatest benefit of DBT would be in breast cancer screening. Following an initial experimental clinical study at Oslo University Hospital Ullevaal from March and May 2010, the Oslo Tomosynthesis Screening Trial will begin in November. Professor Per Skaane MD, who heads the trial, discussed its potential outcome in an interview with Meike Lerne, European Hospital.
An offer to take part in the Oslo Tomosynthesis Screening Trial will be made to all the 50-69-year-old women in Oslo who are invited to Norway’s nationwide breast cancer screening programme, Professor Skaane said. ‘Approximately 16,000 women are invited per year, and with an attendance rate of 60% ,roughly 9,600 women will show up. Out of those 9,600 women we expect 80% -- 7,680 women -- to agree to participate in the tomosynthesis trial.
‘All women will receive conventional 2-D mammography as well as tomosynthesis, which we call 3-D. We’ll have three different arms for independent interpretation of the data: In one arm the radiologists will read conventional 2-D mammography. In the second arm, the reading includes 2-D plus computer-aided detection and, in the third arm, conventional 2-D mammography plus tomosynthesis images will be read.’
All the mammographic examinations are to be independently interpreted by the eight radiologists taking part in the study. ‘There will be a common consensus meeting for all examinations having a positive score in at least one of the study arms. Obviously, we will comply with all relevant European and Norwegian guidelines,’ he pointed out.
Recalls and diagnostic work-up will be according to daily practice at the Oslo University Hospital Ullevaal and the guidelines of the Norwegian Breast Cancer Screening Programme (NBCSP) and, since the project is part of the official NBCSP, interpretations will be carried out on-line into the national database of the NBCSP.
The trial is planned to run for three years. Outcome measures will include the performance indicators for organised screening programmes including recall rate, false positive scores, cancer detection rate, positive predictive values, and cancer characteristics for the individual readers and the study arms of the project. Ultimately, explained Prof. Skaane, ‘We want to find out whether tomosynthesis is better than mammography in detecting breast cancer. The hope is that this new technique not only increases sensitivity, which means it picks up more cancers but also reduces the false positives, meaning it increases specificity. In Europe according to the European guidelines the false positives should be below 3-5 %, this explains the much higher recall rate in the US. We have a much lower recall rate in Europe.’
The women who undergo both mammography and tomosynthesis will be informed about the project, the technique, and the additional compression and radiation dose. What about that double dose of radiation? ‘For tomosynthesis the dose is 1.4 to 1.5 mGy compared to 1.24 mGy for 2D. You have to double it. That means there is a slightly higher exposure, but we discussed this issue with the Norwegian Radiation Protection Authorities,’ the professor explained, adding: ‘Since this is a once in a lifetime dose it was not considered a problem.’
Tomosynthesis advantages
‘The main advantage seems to be in the small distortions of speculate masses. They are much better visualised with tomosynthesis,’ Prof. Skaane pointed out. ‘A challenge will be the time needed for the radiologist to interpret the examinations. Therefore, we intend to record all interpretation times to see if tomosynthesis is cost-effective. Time is money. It is great if we detect more cancers and have a lower recall rate, but if the workload for tomosynthesis is too high we have to question the new technique.’
Use in a clinical setting
‘In March this year, we installed tomosynthesis in the Oslo University Hospital Ullevaal, and we did a pre-test study. Our experience so far is very promising but, as far as I’m concerned, the future of tomosynthesis will be in the screening setting, not in the diagnostic setting. In diagnostics you have high-resolution ultrasound, MRI, biopsy, etc. However, particularly when screening women with dense breast parenchyma, that is women in the younger age group and women on hormone replacement therapy, tomosynthesis might increase sensitivity.
What next?
If tomosynthesis is found to be more effective than mammography, will it be implemented as a screening method? ‘It is too early to say,’ said Prof. Skaane. ‘If it is better, we have to see how much better and how much time is needed to achieve this improved diagnostic performance. There are several factors. A question we have to answer is: Will we need tomosynthesis in both views – craniocaudal and mediolateral oblique – or do we need only one view? Or, should we combine one view of conventional mammography with one tomosynthesis view? For a long time, we discussed whether only women with dense breast parenchyma should be offered screening with this new technique. In the end, we decided to include all women, but we still do not know whether one view is sufficient. Perhaps we will need to do a second study after closing this trial to see how many lesions we missed.’
27.10.2010
